Review on Fourier

Slides edited from:

Prof. Brian L. Evans and Mr. Dogu
Arifler Dept. of Electrical and
Computer Engineering The
University of Texas at Austin course:
EE 313 Linear Systems and Signals
Fall 2003
 Existence of the Fourier Series
Existence T0
f t dt

0

Convergence for all t f t t

Finite number of maxima and minima in one

period of f(t)
 Fourier Integral

G f g t e j 2 f t
dt X xt e j t dt

1
g t G f e j 2 ft
df xt X e j t
d
2

Communicat ion Systems Signal Processing

Conditions for the Fourier transform of g(t) to
exist (Dirichlet conditions):
x(t) is single-valued with finite maxima and minima in
any finite time interval
x(t) is piecewise continuous; i.e., it has a finite number of
discontinuities in any finite time interval

x(t) is absolutely integrable g t dt
 Fourier Transform
What system properties does it possess?
Memoryless
Causal
Linear
Time-invariant
What does it tell you about a signal?
Answer: Measures frequency content
What doesnt it tell you about a signal?
Answer: When those frequencies occurred in time
 Useful Functions
Unit gate function (a.k.a. unit pulse function)
rect(x) 1
1 0 x
2
1
rect x
1
x
x 2 2
-1/2 0 1/2 1 1
x
2
What does rect(x / a) look like?
Unit triangle function
(x) 1
0 x
x
1
2
1
1 2 x x
-1/2 0 1/2
x 2
 Useful Functions
Sinc function
sin x
sinc x
sinc(x) 1

x
How to compute sinc(0)?
x As x 0, numerator and
3 2 0 2 3 denominato r are both going
to 0. How to handle it?
Even function
Zero crossings at x , 2 , 3 , ...
Amplitude decreases proportionally to 1/x
 Fourier Transform Pairs
t jt t / 2 jt
F rect e dt e dt

t t / 2

t t
2 sin sin

1 jt / 2

e jt / 2 2 t 2 t sinc t
e
j t 2

2

f(t) F() t
1 F

-t/2 t/2 t 6 4 2 0 2 4 6
0 t t t t t t
Fourier Transform Pairs
From the sampling property of the impulse,

F t t e j t dt e j 0t 1

f(t) = 1 F() = 2 ()
1 F
(2)

t
0 0

(2) means that the area under the spike is (2)

 Fourier Transform Pairs
1
F 0 1 j 0t

jt
1
e d e
2 2
0

0 or e j 0t 0
1 j 0t
e
2

1 j 0 t
Since cos 0t e e j 0t
2

cos 0t 0 0

f(t) F()
() ()
F
t
0 0 0 0
 Fourier Transform Pairs
1 t 0
sgn t
lim e at u t e at u t
1 t 0 a 0
sgn(t)

F sgn t lim F e at u t F e at u t
a 0
1

1 1
t
lim
a 0 a j -1
a j
2 j 2
lim 2
a 0 a 2
j
Fourier Transform Properties
Fourier vs. Laplace Transform Pairs
f(t) F(s) Region of Convergence F()
e-at u(t) 1 Re{s} > -Re{a} 1
s+a j + a
e-a|t| 2a -Re{a} < Re{s} < Re{a} 2a
a s2 2 + a2
2

(t) 1 complex plane 1

1 2(s) complex plane 2()
u(t) 1 Re{s} > 0 () + 1/(j
s
cos(0t) [( + 0) + ( 0)]
sin(0t) j[( + 0) - ( 0)]
eat u(t) 1 Re{s} > Re{a}
s-a

Assuming that Re{a} > 0

 Duality
Forward/inverse transforms are similar
1
F f t e dt
j t
f t F e j t
d
2
f t F F t 2 f
Example: rect(t/t) t sinc( t / 2)
Apply duality t sinc(t t/2) 2 rect(-/t)
rect() is even t sinc(t t /2) 2 rect(/t)
f(t) t
F()
1

t 6 4 2 0 2 4 6
-t/2 0 t/2 t t t t t t
 Scaling
f t F
Same as Laplace
1
transform scaling property f at F
a a
|a| > 1: compress time axis, expand frequency axis
|a| < 1: expand time axis, compress frequency axis
Effective extent in the time domain is inversely
proportional to extent in the frequency domain
(a.k.a bandwidth).
f(t) is wider spectrum is narrower
f(t) is narrower spectrum is wider
 Time-shifting Property
Shift in time f t t0 e j t0 F
Does not change magnitude of the Fourier transform
Does shift the phase of the Fourier transform by -t0
(so t0 is the slope of the linear phase)
Frequency-shifting Property
e j 0 t f t 2 F 0
e j 0 t f t 2 F 0

cos0t f t F 0 F 0
1
2
cos0t f t F 0 F 0
1 1
2 2
sin 0t f t jF 0 jF 0
1
2

sin 0t f t F 0 F 0
j j
 Modulation
y t f t cos0t
Multiplica tion in the time domain is
convolutio n in the frequency domain :

Y F 0 0
1
2
Recall that

x t t t x t t dt x t


x t t t0 t t0 x t t dt x t t0

So,

Y F 0 F 0
1 1
2 2
 Modulation
F()
Example: y(t) = f(t) cos(0 t) 1

f(t) is an ideal lowpass signal

Assume 1 << 0
-1 0 1
Y()
1/2 F0 1/2
1/2 F0

-0 - 1 -0 + 1 0 0 - 1 0 + 1
0 0
Demodulation is modulation followed by
lowpass filtering
Similar derivation for modulation with sin(0 t)
 Time Differentiation Property
Conditions From the chain rule, recall that
f(t) 0, when |t| u dv u v - v du
f(t) is differentiable Let u e j t and dv df (t ),
Derivation of property: so du j e j t
Given f(t) F()
f t d e j t

B e j t f t


t
d
Let B F f (t ) j

f t e j t dt j F
dt

df t df t
B e j t dt j F
dt dt

e j t df t df n t
n
j n
F
dt
 Time Integration Property
f x dx ?
t
Find
-

From the property of time convolutio n :

f x dx f x u t x dx
t

- -

f t u t
1
F
j
F
F 0
j
Therefore,
F
f x dx F 0
t

- j
 Summary
Definition of Fourier Transform

F f t e j t dt

Two ways to find Fourier Transform

Use definitions
Use properties